The safety and potential risks of many medications used in treating pregnant women and newborn children are not well defined. This is, in part, due to limited knowledge of how readily drugs transfer into integral organs such as the brain early in development. The present Thesis investigated the expression, regulation and function of ATP-binding cassette (ABC) efflux transporters at brain, cerebrospinal fluid (CSF) and placental interfaces of the Sprague Dawley rat, highlighting age-dependent differences. The expression of eight main ABC efflux transporters (abcb1a/b, abcg2, abcc1-5) was described in the rat brain, choroid plexus and placenta, revealing distinct developmental profiles. Immunostaining of the interface between the ventricular CSF and the brain identified PGP and BCRP transporters at this interface in the newborn rat but not in the adult. Drugs (rhodamine-123, digoxin, cimetidine, paracetamol) entered the developing brain more than the adult brain. Chronic drug exposure (diallyl sulfide, digoxin, paracetamol) caused an up-regulation of efflux transporter expression and functional capacity at brain barriers in adult but not fetal (E19) or newborn (P4) rats. Fetal rats exposed chronically to paracetamol via the dam had increased transfer into the brain compared to acute treatment, with this result being dose-dependent. Transcriptomic analysis revealed that chronic paracetamol exposure caused a large inflammatory profile in the placenta, indicating potential toxicity of paracetamol use in pregnancy over extended periods. The results in this Thesis suggest that drug transport into the rat brain is higher during development for acute and chronic conditions and indicate aspects of a treatment regime that may need to be considered differently for the safe prescription of medications to patients of different ages.

Despite current advances in therapies and the gradual decline in breast cancer-related mortality, metastasis is the major determinant of breast cancer survival. The treatment regimens for breast cancer metastasis are complicated by the unpredictability of metastasis development, as well as the inter- and intra-tumour heterogeneity observed in breast cancer patients. Therefore, the diagnosis and management of metastatic disease remains a major therapeutic challenge for breast cancer treatment. To overcome these obstacles, full understanding of the molecular mechanism that governs the process of metastasis is urgently needed. Annexin A1 is a protein well known for its anti-inflammatory biology and also demonstrated convincing, though controversial influences on breast cancer progression. Accumulating evidence suggest that the influence of annexin A1 may only manifest in the most aggressive breast tumour and investigation of this relationship needs to be conducted in models that reflect this specificity A powerful model developed to effectively study the complexities associated with breast cancer metastasis are breast cancer variant cell lines derived from the same parental line but displaying differing in metastatic capabilities. The MDA-MB-231HM.LNm5 is a one of such novel cell lines derived by in vivo passaging of the TN human breast adenocarcinoma MDA-MB-231 line and demonstrated robust metastatic propensity. Using this cellular model of metastasis, the body of work presented in this thesis attempted to investigate the mechanisms underlying the acquisition of metastatic phenotype in both a hypothesis and a non-hypothesis-driven manner. Part one of this thesis aimed to describe some of the phenotypic, molecular, and transcriptomic changes associated with enhanced metastatic potential found in the MDA-MB-231HM.LNm5 line, including changes in energy metabolism, proliferation, and growth-related processes. In the second part of this thesis, the endogenous expression of annexin A1 in the MDA-MB-231HM.LNm5 cells line and the parental line was silenced to explore the influence of annexin A1 on characteristics associated with elevated metastatic potential. Finally, by manipulating tumour and host ANXA1 levels, as well the host immune system, we attempted to unravel the involvement of ANXA1 in tumour initiation, growth and metastasis in vivo. Accompanying the aggressive in vivo metastasis phenotype, the MDA-MB-231HM.LNm5 metastatic daughter line exhibits heightened energy metabolism and chemo-resistance but reduced in vitro proliferative propensity. This ‘go or grow’ dichotomy, underlined by an increase in the quiescent cell population and dampened Ca2+ signalling, can be restored by the knock-down of ANXA1. In comparison, phenotypes of the non-metastatic parental MDA-MB-231 cells are unaltered by ANXA1 KD. In vivo studies showed that the expression of annexin A1 is required for growth and progression of breast cancer in specific metastatic mouse models. Overall, the studies presented in this thesis deepens our knowledge of the complex biological processes underlying the acquisition of metastatic propensity in breast cancer and reported evidence that suggested annexin A1 to be an important contributor of metastatic alterations. Further targeted investigation could facilitate the development of new strategies for therapeutic interventions and clinical management of patients with metastatic breast cancer

Neurodegenerative diseases such as Frontotemporal Dementia (FTD), and Amyotrophic Lateral Sclerosis (ALS), also known as Motor Neuron Disease (MND), have limited therapeutics available, placing a high demand on healthcare systems and overall economic and social burden. Recently, it has been found that pathological mutations in RNA binding proteins such as TDP-43 (Tar DNA-binding Protein, 43kDa), FUS (Fused in Sarcoma), and hnRNP (heterogenous nuclear RiboNucleoProteins) can be causative of ALS or FTD. Cell stress has been implicated in ALS and FTD, particularly oxidative stress, mitochondrial dysfunction, ATP depletion and metal dyshomeostasis. In cell culture models, it has been shown that TDP-43 and hnRNP proteins can incorporate into stress granules in conditions of cell stress, including during oxidative stress, metal dyshomeostasis or conditions of ATP depletion, however this is not well defined. Furthermore, there is limited evidence of FUS incorporating within stress granules. The study described here has enhanced the understanding of RNA binding proteins in vitro. Utilising neurodegeneration-associated cell stress pathways in conjunction with advanced bioimaging techniques, we show that both non-mutated FUS and non-mutated TDP-43 can occur within the same stress granule. Furthermore, we show different populations of TDP-43 and FUS within stress granules from different stress types. This study also enhanced our understanding of the sub-cellular distribution of a neuroprotective metal-complex therapeutic, utilising fluorescence lifetime imaging on live cells; this therapeutic has previously been shown to be neuroprotective against TDP-43 aggregation. To extend the study, a novel metal complex was utilised in attempt to enhance the field of techniques available to study RNA binding proteins in vitro. Overall, this thesis has advanced our understanding of disease pathways, therapeutic action, and new investigative tools for neurodegeneration.

Alzheimer’s disease (AD) is a neurodegenerative disease that is pathologically characterized by abnormal deposition of extracellular amyloid plaques and intraneuronal neurofibrillary tangles. The deposition of these aggregated proteins causes progressive brain atrophy resulting from gradual synaptic loss and neuronal cell death. Although the aetiology of AD remains elusive, studies have shown that amyloid beta (Aβ) peptide, a cleavage product from amyloid precursor protein (APP), is a key protein causing AD pathogenesis. Recent studies have identified how the presence of soluble low molecular weight Aβ oligomers in the brain correlate best with synaptic loss, and they are a better predictor of disease progression compared to the presence of amyloid plaques or neurofibrillary tangles. In the AD brain, neuronal subpopulations appear to exhibit different levels of vulnerability to Aβ, particularly the basal forebrain cholinergic and hippocampal glutamatergic neurons, while GABAergic neurons appear to remain unaffected till later disease stages. Current treatments based on knowledge gathered from mouse models targeting the cholinergic and glutamatergic systems only alleviate symptoms and are ineffective in halting disease progression. Therefore, we hypothesize that Aβ exerts its neurotoxic effect by binding to a subpopulation of mature neurons. To address this, human embryonic stem cells (hESCs) were differentiated into mature glutamatergic and GABAergic neurons and cultured up to 12 weeks. These cultures were treated fortnightly with soluble synthetic Aβ peptide for 96 hrs. We found that Aβ bound to neurites in culture, altered gene expression and neurotoxicity was more pronounced in 6-week old glutamatergic than GABAergic cultures. Further investigations in determining the specific toxic species of Aβ oligomers revealed that 12 week old cultures were more susceptible to Aβ oligomer induced toxicity, with Aβ dimers being most toxic to glutamatergic neurons while GABAergic neurons were most susceptible to Aβ tetramers. In summary, we successfully established a simple hESC based model to study Aβ toxicity. Our findings also highlight the importance of using relevant human cell-based models to study AD pathogenesis as well as identify potential AD modifying therapeutic strategies.

Motor neurone disease (MND) is a fatal human neurodegenerative disorder. The most common form of MND is amyotrophic lateral sclerosis (ALS). MND is characterised by the progressive destruction of motor neurons in the central nervous system which causes muscle weakness, muscle atrophy, paralysis and ultimately death. The sporadic forms of the disease account for the majority of patients, and 5-10% of MND cases are inherited (familial MND) (Marin et al., 2017). Both sporadic and familial MND share similar clinical and pathological features, suggesting common molecular mechanisms of degeneration. Among the familial MND patients approximately 20% possess a mutation in the SOD1 gene encoding for the enzyme Cu/Zn superoxide dismutase (Rosen et al., 1993). There are more than 170 different SOD1 gene mutations described, and the majority are missense substitutions resulting in a toxic gain of enzyme function (http://alsod.iop.kcl.ac.uk/). Transgenic mouse models over-expressing mutant forms of the human SOD1 gene replicate key pathological symptoms seen in MND patients and are widely used to study MND. Despite progress in deciphering the molecular mechanisms of this disease, the cause and modulation of MND remains unclear. The Amyloid Precursor Protein (APP), is well-known for its association with Alzheimer's Disease, and it has been shown to be a modulator of MND. APP protein expression levels were increased in the spinal cords from MND patients as well as in SOD1 transgenic mice at symptomatic stage of the disease (Koistinen et al., 2006; Rabinovich-Toidman et al., 2015). The resultant SOD1-G93A:APP-/- mice from the cross breeding between APP homozygous deletion and SOD1-G93A transgenic mice (overexpress human SOD1 gene with G93A familial mutation) showed significant decrease in MND pathogenesis and reduced disease progression (Bryson et al., 2012). The SOD1-G93A:APP-/- mice also displayed significantly ameliorated muscle contractility, improved neuromuscular junction innervation and decreased motor neuron loss. Taken together these findings suggest an important role for APP in MND pathophysiology. APP is part of a gene family that includes the amyloid precursor-like protein 1 (APLP1) and amyloid precursor-like protein 2 (APLP2) genes. To understand if other APP-family members modulated MND we investigated the role of APLP2 in the SOD1-G37R transgenic mouse model. We found a significant sex-dependent increase in the expression of APLP2 protein in the spinal cord of the SOD1-G37R mice. To test if APLP2 gene expression can modulate disease outcomes in MND we crossed the SOD1-G37R and APLP2 knockout (KO) mice to generate the SOD1:APLP2+/- and SOD1:APLP2-/- lines. We found the lack of APLP2 expression improved motor performance and extend survival in a sex-dependent manner. The molecular basis for APLP2’s actions identified effects on muscle physiology and synaptic function at the neuromuscular junction. Taken together, our novel results demonstrate there are sex-dependent differences in the SOD1 mouse model, and this is affected by APLP2 expression. These data extend the modulatory role by the amyloid precursor protein family in MND, and identify the APP-family as an important target for further investigation into the cause and regulation of MND.

Amyloid precursor protein (APP) is neuroprotective in traumatic brain injury (TBI). Treatment with soluble amyloid precursor protein (sAPP) can rescue motor and cognitive deficits following TBI in mouse and rat models (Corrigan et al., 2012c). The neuroprotective active site in sAPP is located in residues 96 to 110 (APP96-110) (Corrigan et al., 2014). We hypothesize that APP96-110 interacts with a specific molecule(s) to trigger its neuroprotective response in TBI. To identify protein(s) interacting with the APP96-110 peptide, a biotin-streptavidin affinity capture method combined with mass spectrometry was utilised. Among the proteins identified, the Amyloid Precursor-like Protein 2 (APLP2) was found to be a robust interacting target for APP96-110. Previous reports showed APLP2 binds to a region in APP which includes 96-110 (Soba et al., 2005). To test the role of APLP2 in TBI, APLP2 wildtype (APLP2+/+) and APLP2 knockout (APLP2-/-) mice, from both sexes, were subjected to mild controlled cortical impact injury. Brains were collected following 7 days of surgery and histopathological assessment was done looking at primary and secondary effects of injury. These include tissue morphology, neuronal loss, axonal injury, tau pathology, astrogliosis and microgliosis. Motor function was assessed by DigiGait over 7 days post-surgery. Initial gait analysis showed the craniotomy procedure itself induced gait disturbances, but to a lesser extent and that injury worsened gait performance in both genotypes. Sex differences were observed in brain injury, with males more susceptible at acute phase of injury with increased motor deficits and astrogliosis in both genotypes. There was greater axonal damage and tau pathology detected in males than females expressing endogenous APLP2. This study highlights the importance of considering craniotomy controls and female and male mice in TBI study. Sex-specific comparisons made between APLP2+/+ and APLP2-/- mice following injury showed the lack of APLP2 in males leads to decreased motor deficits, axonal damage and tau pathology compared to males expressing endogenous APLP2. In the case of females, APLP2-/- mice were less susceptibility to brain injury compared to APLP2+/+ females. This suggests APLP2 expression may be modulated by sex hormones. Using an in silico approach, progesterone and estrogen transcription binding motifs were identified in the mouse and human APLP2 promoter sequence.

Fibrosis is one of the leading causes of death that may affect all organs in the human body. All forms of fibrosis are characterized by fibrotic scarring resulting from replacement of normal tissue with extracellular matrix (ECM). Idiopathic pulmonary fibrosis (IPF) is the most common form of idiopathic interstitial pneumonias; it is relentlessly progressive and ultimately fatal. The aberrant accumulation of myofibroblasts and the excessive deposition of ECM are the main features of IPF. The suggested sources of myofibroblasts include resident lung fibroblasts, lipofibroblasts, epithelial-mesenchymal transition (EMT), endothelial-mesenchymal transition (EndoMT) and recruitment from circulating fibrocytes. Lipofibroblasts are lipid-droplet containing cells, which have a key role in maintaining homeostasis of the lung. However, IPF patients have stiffer lungs and homeostatically dysregulated cellular microenvironments. Stiff substrates are known to augment myofibroblast differentiation, ECM production and the activation of the prototype profibrotic cytokine, transforming growth factor β1 (TGF-β1). In contrast, stiff substrates inhibit cyclooxygenase-2 (COX-2) expression and the synthesis of the anti-fibrotic prostanoid, prostaglandin E2 (PGE2). This thesis for the first time revealed a reduction in the lipofibroblast population and prostaglandin E synthase (PTGES) level in the lungs of IPF patients. This current study also highlighted the significance of dimensionality in modulating fibroblast behavior by establishing human lung fibroblasts in 2D soft and 3D soft microenvironments (that have similar stiffness) and performing head-to-head comparisons with the conventional 2D stiff cultures. Our data demonstrated the marked suppression of fibroblast proliferation and fibrogenesis in 3D soft microenvironment, as contrasted with 2D soft, independent of differences in prostanoid levels. This thesis also presented evidence that suggested the augmented anti-fibrogenic actions of glucocorticoid and PGE2 combination. Within this thesis, I showed that the phenotypic plasticity of fibroblasts depends on substrate stiffness and dimensionality. The level of lipid-droplet inclusions that mark the formation of lipofibroblasts was remarkably increased in 3D fibroblasts. In 2D soft and 3D microenvironments, striking reductions of the myofibroblast marker, alpha smooth muscle actin (ACTA2) and increases in the lipofibroblast markers, adipose differentiation-related protein (ADRP) were observed. Activated exogenous TGF-β1, despite elevating expression of the fibrogen interleukin-11 (IL-11), did not induce the differentiation of lipofibroblast into myofibroblasts in 2D soft 3D soft settings. This study has not only provided insight into fibroblast phenotypic plasticity, but has also revealed a novel role of parathyroid hormone-related protein (PTHrP), as a main effector in mediating the effects of softness or dimensionality on fibroblast phenotypic plasticity. Moreover, this study has established the anti-fibrogenic actions of PTHrP in lung fibroblasts in vitro.

Traumatic brain injury (TBI) represents a major cause of disability and death worldwide with sustained neuro-inflammation and autophagy dysfunction contributes to cellular damage. Stimulator of interferon genes (STING)-induced type-I interferon (IFN) signaling is known to be essential in mounting the innate immune response against infections and cell injury in the periphery, but its role in the CNS remains unclear. Recent studies also implicated the STING and type-I interferon pathways in autophagy activation. We have previously identified the type-I IFN pathway as a key mediator of neuro-inflammation and neuronal cell death in TBI (Karve et al., 2016). This thesis has explored the modulation of the type-I IFN and neuro-inflammatory responses by STING and its contribution to neuronal cell death and autophagy activity after TBI in vivo. Additionally, a role for STING in regulating autophagy activity and cellular viability in response to H2O2-induced oxidative stress has also been investigated in vitro. To investigate this, C57BL/6J wildtype (WT) and STING-/- mice (8-10-week-old male) were subjected to controlled-cortical impact (CCI) surgery. It was found that STING expression was upregulated by 1.24 + 0.20 and 4.45 + 0.93 fold at 2h and 24h, respectively in WT mice after CCI as determined by qPCR with increased expression confirmed by Western blot and immunohistochemistry. This correlated with an elevated pro-inflammatory cytokine profile with an upregulation in TNF-α, IL-1β and type-I IFN (IFN-α and IFN-β) levels and heightened glial reactivity. Significantly, this expression was suppressed in the STING-/- mice with a smaller lesion volume in the knockout animals at 24h post CCI (WT = 4.16 + 0.27mm3, STING-/- = 3.20 + 0.17 mm3; p< 0.05) as assessed by triphenyl tetrazolium chloride (TTC) staining. Supporting a role for STING in human TBI, a significant upregulation in STING expression (2.25 + 0.50 fold; p< 0.0001) was detected in late trauma human brain samples as compared to the control group. Further, impaired autophagy activation with concurrent increased levels of LC3-II, p62 and LAMP2 were detected in WT mice 24h after TBI. However, STING-/- mice showed reduced LAMP2 expression as compared with LC3-II and p62 levels 24h after TBI suggesting a role for STING in driving dysfunctional autophagy seen in TBI. Interestingly, ablation of the STING pathway in vitro revealed a differential role for STING in promoting cellular survival following H2O2-induced oxidative stress evidenced by higher cellular viability in the WT MEF SV40 cells as compared to STING-/- MEF SV40 cells line. Supporting a role for STING in promoting cellular survival after H2O2 treatment, it was found that STING-/- MEF SV40 cells show impaired autophagy activity as compared to the WT cells following H2O2 insult. This thesis uncovers a novel role for STING-mediated type-I IFN signalling in regulating neuro- inflammatory processes and autophagy activity after TBI. More importantly, this thesis demonstrates a multi-faceted role for STING in the CCI animal model of TBI and in the cell- based model of oxidative stress induced by H2O2. Taken together, these findings implicate for the first time, a detrimental role for STING in mediating the TBI-induced neuro- inflammatory response and autophagy dysfunction and has potentially identified a new therapeutic target for reducing the cellular damage in TBI.

AD is a complex disease, involving the perturbation of multiple interrelated biological pathways. Despite being the most common form of neurodegenerative disease, with the number of patients expected to increase drastically, the aetiology of AD remains unknown. Although genetic factors such as mutations in the APP and PSEN genes, which drive the overproduction of Aβ, are known to cause early on-set familial AD (EOFAD), majority of AD patients (> 95%) who develop clinical symptoms of AD later in life (i.e. late on-set AD (LOAD)) do not carry these mutations. Contrary to earlier assumptions that occurrence of LOAD was sporadic, GWAS and exome sequencing studies have identified genetic variants, in genes that are microglia specific, or highly expressed in microglia, that are associated with increased risk of LOAD, suggesting altered regulation of microglial functions to be involved in the pathogenesis of LOAD. Using microglia isolated from the 5xFAD mouse model of AD, we identified by bulk and single cell RNA-seq that plaque phagocytosing (X04+) and non-plaque phagocytosing (X04-) microglia are distinct microglial populations separable by their transcriptomic signature. Our study suggests that X04+ microglia are a homogenous population of microglia with a distinct gene expression profile associated with amyloid uptake. In contrary, X04- microglia are associated with an age-related transcriptomic profile and may be involved in the over-pruning of synapses in 5xFAD mice. We also demonstrated that microglia can undergo transcriptomic changes upon exposure to different environmental cues. Although therapeutic approaches that target different pathological features of AD have been evaluated, clinical translation of therapeutics has been very unsuccessful. This could be in part due to potential therapeutics targeting a single disease pathology. To identify new therapeutic candidates for the treatment of AD, this study also explored the use of metal-based compounds as a multi-targeting therapeutic strategy. Five novel metal-based compounds were screened to identify leading compounds that confer neuroprotective, metal-regulating and inflammation-modulating effects in a generic model of neuroinflammation. LM47 was identified as the only leading compound that exerted inflammatory-modulating activity through copper-associated action. Further testing of LM47 showed that the compound was well-tolerated in vivo. However, data by pharmacokinetic study and ICP-MS suggest that the copper free ligand LM46, and not LM47, acts as the active compound in vivo. Treatment of 5xFAD mice with LM46 increased the proportion of X04+ microglia in brain. However, Both LM46 and LM47 treatment of 5xFAD mice induced increased brain Aβ plaque load in the animals. Taken together, findings from our study suggest that improved targeting of specific microglia sub-population in AD could confer therapeutic outcomes.

The airway epithelium is an important barrier interface that protects the lungs against environmental insults and pathogens. Airway epithelial cells are increasingly recognized as key regulators of lung immune homeostasis by orchestrating various aspects of both innate and adaptive immunity. Dysfunction of airway epithelial cell functions have been strongly associated with the pathogenesis of various airway disorders. In this body of work, I identify the expression of a novel membrane-spanning protein, MS4A8B, on the motile cilia of airway epithelial cells. Predictive protein sequence analysis reveals the presence of an immunoreceptor tyrosine-based inhibitory motif (ITIM) in its intracellular C- terminal domain. MS4A8B is a member of the MS4A protein family, which comprises a group of 4-domain membrane-spanning proteins. Whilst the core functions of MS4A proteins have not been established, they have been suggested to play active roles in cellular signaling, cellular proliferation and differentiation. Examination of MS4A8B expression in the human airways reveals a highly restricted pattern of expression that is induced only upon mucociliary differentiation of airway epithelial cells. MS4A8B protein is localized to the motile cilia of the airway epithelium in human airway biopsies and in differentiated primary bronchial airway epithelial cell (PBEC) cultures. The expression of MS4A8B decreased in various airway inflammatory disorders including asthma and chronic obstructive pulmonary disease (COPD). In vitro assessment of MS4A8B function reveal that it possesses immune-regulatory functions, with the putative ITIM motif predicted to be a primary effector. In vitro phosphorylation studies show that the putative ITIM motif is capable of being tyrosine phosphorylated, with site-directed mutagenesis of a key tyrosine residue contained in the motif significantly impairing this ability. Combined findings suggest that MS4A8B may exert an important ‘braking’ mechanism on proinflammatory cytokine production in airway epithelial cells, with its observed loss in various airway disorders likely resulting in dysregulated expression of proinflammatory cytokines and the exacerbation of disease symptoms.

Glucocorticoids (GCs) remain the frontline treatment in the management of chronic inflammatory diseases, as they are the most potent and effective anti-inflammatory agents available so far. However, impaired responses to glucocorticoid therapy in some patients with severe disease remain a challenging clinical problem. The airway epithelial function influences inflammation in chronic respiratory diseases. Epithelium, as the site of deposition of inhaled glucocorticoids (ICS), is a key target of GC action. Synthetic GCs, including ICS, exert anti-inflammatory effects in airway epithelium by transactivation of genes and by inhibition of release of pro-inflammatory cytokines. Emerging evidence suggests that physiological GC, cortisol, might act as a partial agonist at the glucocorticoid receptor (GR) in the airway epithelium. However, whether cortisol can be a limiting factor to beneficial effects of synthetic GCs, remains to be established. Therefore, through a better understanding of the impact of cortisol on the effects of synthetic GCs in vitro and in vivo, as well as of novel individual mediators of GC actions in the airway epithelium, new strategies may arise for restoring GC responsiveness. Data presented within this thesis has provided evidence that cortisol acts like a partial agonist at the glucocorticoid receptor, limiting GC-induced GC Receptor-dependent transcription in the BEAS-2B human bronchial epithelial cell line. Cortisol also limited the inhibition of granulocyte macrophage colony-stimulating factor (GM-CSF) release by synthetic GCs in TNFα-activated BEAS-2B cells. The relevance of these findings is supported by observations on tracheal epithelium obtained from mice treated for 5 days with systemic GC, showing limitations in selected GC effects, including inhibition of pro-inflammatory cytokine IL-6. Moreover, gene transactivation by synthetic GCs was compromised by standard air-liquid interface (ALI) growth medium cortisol concentration of 1.4 µM in the ALI differentiated organotypic culture of primary human airway epithelial cells. These findings suggest that endogenous corticosteroids may limit certain actions of synthetic pharmacological GCs and contribute to GC insensitivity, particularly when corticosteroid levels are elevated by stress. Data obtained during these thesis studies also highlight the potential of the transcriptional repressor, promyelocytic leukaemia zinc finger (PLZF) to mediate selected glucocorticoid effects in the airway epithelium, including the induction of targets important in mediating physiological effects on the normal lung development and of ones with relevance to the distinct glucocorticoid effects on the epithelial restitution following inflammation and injury. This thesis has provided novel insights into mechanisms of glucocorticoid action and insensitivity in the airway epithelium, allowing the development of strategies for improved treatment of chronic airway inflammatory diseases.

Understanding transition metal dys/homeostasis holds potential for the discovery of new treatments to curb the increasing burden of cardio- and cerebrovascular diseases. However, their pharmacology and roles in the cardiovascular system have been underexplored. This thesis investigated examples of transition metal pharmacology in the cardiovascular system by (1) using genetically-modified mice susceptible to iron accumulation and (2) pharmacologically elevating the levels of zinc in isolated arteries to probe its roles. To understand the role of iron, two age groups of mice (13 mo and 23 mo) deficient in the microtubule-associated protein tau (MAPT, whose role in iron homeostasis of the brain was recently established) and their wild type counterparts were used. Loss of tau protein presented an accelerated age-dependent cardiomyopathy phenotype accompanied by cardiac iron accumulation. Increased systolic blood pressure, cardiac hypertrophy and lower basal right atrial rate were seen as early as 13 mo and were further aggravated by 23 mo. Notably, 23 mo wild type mice also showed a similar phenotype with a significant iron, zinc and copper accumulation together with an altered sensitivity of isolated mesenteric arteries to contractile agonists. More importantly, chronic treatment of the middle-aged tau KO mice with clioquinol prevented the increase in systolic blood pressure, cardiac hypertrophy and lowering of the basal right atrial rate. Clioquinol is a moderate Fe2+ chelator and a Zn2+ ionophore. Thus, by using compounds with similar properties (ionophores that deliver metals across membrane and chelators that remove them), the role of zinc in vascular tone was explored in rat isolated small resistance arteries. Four different zinc ionophores from structurally-distinct chemical classes all caused a concentration-dependent relaxation of mesenteric arteries pre-contracted by a range of agonists. The potency and efficacy were comparable to the established vasodilator drugs, sodium nitroprusside and acetylcholine. The effects were also observed in middle cerebral, basilar, coronary, saphenous and pulmonary arteries. The archetypal ionophore Zn(DTSM) also showed a depressor effect in vivo. Importantly, removal of the endogenous zinc from resting cerebral and coronary arteries with basal tone, but not the others, caused an active contraction suggesting a crucial role of zinc in maintaining resting vascular tone. In isolated small mesenteric arteries, the mechanism of zinc ionophore-induced vasorelaxation was then investigated. Vasorelaxation by zinc ionophores involved a combination of non-competitive inhibition of voltage-operated Ca2+ channels (common to all the four classes) and a competitive α1-adrenoceptor antagonism (clioquinol). The release of vasodilatory mediators from the endothelium or perivascular nerves, antagonising the actions of endothelin, arginine vasopressin and thromboxane A2 receptors, activating potassium channels (voltage-operated, ATP-sensitive and calcium-activated) were all not involved in the relaxation. This study revealed an essential physiological role tau protein in cardiovascular function and for the first time reported a novel vasodilatory property of zinc ionophores. These results confirm that a more detailed investigation of zinc and other transition metals in the cardiovascular system holds untapped potential for the discovery of novel biology, treatment targets and therapies.